Recommender System for Telecom Product and Services

This paper presents the development and implementation of a personalized recommender system for Etisalat's product and services department, specifically targeting internet data bundles. By leveraging extensive customer data, including demographic information, subscription details, internet usage patterns, and customer behavior, the system aims to provide highly accurate recommendations that align closely with individual customer interests. The recommender system is built using TensorFlow, an opensource machine learning framework, to ensure robust performance and scalability. The main goal is to keep the existing customer satisfied and acquire new customers and gain market competence in hand. Our results demonstrate significant improvements in recommendation precision and customer satisfaction, highlighting the potential of machine learning in enhancing customer experience in the telecom industry.